When a human sees an object, certain neurons in our brain's visual cortex light up with activity, but when we take hallucinogenic drugs, these drugs overwhelm our serotonin receptors and lead to the distorted visual perception of colours and shapes. Similarly, deep neural networks that are modelled on structures in our brain, stores data in huge tables of numeric coefficients, which defy direct human comprehension. But when these neural network's activation is overstimulated (virtual drugs), we get phenomenons like neural dreams and neural hallucinations. Dreams are the mental conjectures that are produced by our brain when the perceptual apparatus shuts down, whereas hallucinations are produced when this perceptual apparatus becomes hyperactive. In this blog, we will discuss how this phenomenon of hallucination in neural networks can be utilized to perform the task of image inpainting. Image inpainting is the art of synthesizing alternative contents for the reconstruction of missing or deteriorated parts of an image such that the modification is semantically correct and visually realistic.

Neural Network uses optimising strategies like stochastic gradient descent to minimize the error in the algorithm. The way we actually compute this error is by using a Loss Function. It is used to quantify how good or bad the model is performing. These are divided into two categories i.e.Regression loss and Classification Loss. In this article, we will cover some of the loss functions used in deep learning and implement each one of them by using Keras and python.

Classification of polarimetric synthetic aperture radar (PolSAR) images is an active research area with a major role in environmental applications. The traditional Machine Learning (ML) methods proposed in this domain generally focus on utilizing highly discriminative features to improve the classification performance, but this task is complicated by the well-known "curse of dimensionality" phenomena. Other approaches based on deep Convolutional Neural Networks (CNNs) have certain limitations and drawbacks, such as high computational complexity, an unfeasibly large training set with ground-truth labels, and special hardware requirements. In this work, to address the limitations of traditional ML and deep CNN based methods, a novel and systematic classification framework is proposed for the classification of PolSAR images, based on a compact and adaptive implementation of CNNs using a sliding-window classification approach. The proposed approach has three advantages. First, there is no requirement for an extensive feature extraction process. Second, it is computationally efficient due to utilized compact configurations. In particular, the proposed compact and adaptive CNN model is designed to achieve the maximum classification accuracy with minimum training and computational complexity. This is of considerable importance considering the high costs involved in labelling in PolSAR classification. Finally, the proposed approach can perform classification using smaller window sizes than deep CNNs. Experimental evaluations have been performed over the most commonly-used four benchmark PolSAR images: AIRSAR L-Band and RADARSAT-2 C-Band data of San Francisco Bay and Flevoland areas. Accordingly, the best obtained overall accuracies range between 92.33 - 99.39% for these benchmark study sites.

Sleep state classification is vital in managing and understanding sleep patterns and is generally the first step in identifying acute or chronic sleep disorders. However, it is essential to do this without affecting the natural environment or conditions of the subject during their sleep. Techniques such as Polysomnography(PSG) are obtrusive and are not convenient for regular sleep monitoring. Fortunately, The rise of novel technologies and advanced computing has given a recent resurgence to monitoring sleep techniques. One such contactless and unobtrusive monitoring technique is Ballistocradiography(BCG), in which vitals are monitored by measuring the body's reaction to the cardiac ejection of blood. In this study, we propose a Multi-Head 1D-Convolution based Deep Neural Network to classify sleep-wake state and predict sleep-wake time accurately using the signals coming from a BCG sensor. Our method achieves a sleep-wake classification score of 95.5%, which is on par with researches based on the PSG system. We further conducted two independent studies in a controlled and uncontrolled environment to test the sleep-wake prediction accuracy. We achieve a score of 94.16% in a controlled environment on 115 subjects and 94.90% in an uncontrolled environment on 350 subjects. The high accuracy and contactless nature of the proposed system make it a convenient method for long term monitoring of sleep states.

In an effort to improve generalization in deep learning, we propose SALR: a sharpness-aware learning rate update technique designed to recover flat minimizers. Our method dynamically updates the learning rate of gradient-based optimizers based on the local sharpness of the loss function. This allows optimizers to automatically increase learning rates at sharp valleys to increase the chance of escaping them. We demonstrate the effectiveness of SALR when adopted by various algorithms over a broad range of networks. Our experiments indicate that SALR improves generalization, converges faster, and drives solutions to significantly flatter regions.

Modern deep learning is primarily an experimental science, in which empirical advances occasionally come at the expense of probabilistic rigor. Here we focus on one such example; namely the use of the categorical cross-entropy loss to model data that is not strictly categorical, but rather takes values on the simplex. This practice is standard in neural network architectures with label smoothing and actor-mimic reinforcement learning, amongst others. Drawing on the recently discovered continuous-categorical distribution, we propose probabilistically-inspired alternatives to these models, providing an approach that is more principled and theoretically appealing. Through careful experimentation, including an ablation study, we identify the potential for outperformance in these models, thereby highlighting the importance of a proper probabilistic treatment, as well as illustrating some of the failure modes thereof.

Deep neural networks are vulnerable to evasion attacks, i.e., carefully crafted examples designed to fool a model at test time. Attacks that successfully evade an ensemble of models can transfer to other independently trained models, which proves useful in black-box settings. Unfortunately, these methods involve heavy computation costs to train the models forming the ensemble. To overcome this, we propose a new method to generate transferable adversarial examples efficiently. Inspired by Bayesian deep learning, our method builds such ensembles by sampling from the posterior distribution of neural network weights during a single training process. Experiments on CIFAR-10 show that our approach improves the transfer rates significantly at equal or even lower computation costs. Intra-architecture transfer rate is increased by 23% compared to classical ensemble-based attacks, while requiring 4 times less training epochs. In the inter-architecture case, we show that we can combine our method with ensemble-based attacks to increase their transfer rate by up to 15% with constant training computational cost.

We previously proposed a method that allows for nonparallel voice conversion (VC) by using a variant of generative adversarial networks (GANs) called StarGAN. The main features of our method, called StarGAN-VC, are as follows: First, it requires no parallel utterances, transcriptions, or time alignment procedures for speech generator training. Second, it can simultaneously learn mappings across multiple domains using a single generator network and thus fully exploit available training data collected from multiple domains to capture latent features that are common to all the domains. Third, it can generate converted speech signals quickly enough to allow real-time implementations and requires only several minutes of training examples to generate reasonably realistic-sounding speech. In this paper, we describe three formulations of StarGAN, including a newly introduced novel StarGAN variant called "Augmented classifier StarGAN (A-StarGAN)", and compare them in a nonparallel VC task. We also compare them with several baseline methods.

Neural networks are vulnerable to input perturbations such as additive noise and adversarial attacks. In contrast, human perception is much more robust to such perturbations. The Bayesian brain hypothesis states that human brains use an internal generative model to update the posterior beliefs of the sensory input. This mechanism can be interpreted as a form of self-consistency between the maximum a posteriori (MAP) estimation of an internal generative model and the external environment. Inspired by such hypothesis, we enforce self-consistency in neural networks by incorporating generative recurrent feedback. We instantiate this design on convolutional neural networks (CNNs). The proposed framework, termed Convolutional Neural Networks with Feedback (CNN-F), introduces a generative feedback with latent variables to existing CNN architectures, where consistent predictions are made through alternating MAP inference under a Bayesian framework. In the experiments, CNN-F shows considerably improved adversarial robustness over conventional feedforward CNNs on standard benchmarks.

We focus on prediction problems with high-dimensional outputs that are subject to output validity constraints, e.g. a pseudocode-to-code translation task where the code must compile. For these problems, labeled input-output pairs are expensive to obtain, but "unlabeled" outputs, i.e. outputs without corresponding inputs, are freely available and provide information about output validity (e.g. code on GitHub). In this paper, we present predict-and-denoise, a framework that can leverage unlabeled outputs. Specifically, we first train a denoiser to map possibly invalid outputs to valid outputs using synthetic perturbations of the unlabeled outputs. Second, we train a predictor composed with this fixed denoiser. We show theoretically that for a family of functions with a high-dimensional discrete valid output space, composing with a denoiser reduces the complexity of a 2-layer ReLU network needed to represent the function and that this complexity gap can be arbitrarily large. We evaluate the framework empirically on several datasets, including image generation from attributes and pseudocode-to-code translation. On the SPoC pseudocode-to-code dataset, our framework improves the proportion of code outputs that pass all test cases by 3-5% over a baseline Transformer.